Batch mode reinforcement learning based on the synthesis of artificial trajectories.
Raphael FonteneauSusan A. MurphyLouis WehenkelDamien ErnstPublished in: Ann. Oper. Res. (2013)
Keyphrases
- batch mode
- reinforcement learning
- incremental learning
- active learning
- control policy
- batch mode active learning
- supervised learning
- learning algorithm
- semi supervised
- state space
- computationally expensive
- online algorithms
- machine learning
- multiple instance learning
- data sets
- model selection
- learning problems
- support vector
- objective function